© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 89876
    Prev
    Next
    INDEX
    Explanations

    French cities and places

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron fires strongly on capitalized tokens that are parts of multiword proper names (especially place names).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Parisian
    -0.96
    sterreich
    -0.85
     Indiana
    -0.84
    lder
    -0.80
     Duque
    -0.80
    épendance
    -0.80
     tačiau
    -0.76
     preponder
    -0.75
     Decreto
    -0.75
    rendre
    -0.73
    POSITIVE LOGITS
     Toulouse
    1.05
     Occ
    1.01
     toul
    0.96
     consuls
    0.92
     Toul
    0.92
     Carcass
    0.91
     Pyrene
    0.89
     Méditerranée
    0.88
     Andorra
    0.87
     Sorg
    0.82
    Activations Density 0.030%

    No Known Activations